ASSERT extracts legacy IT security concepts into document graphs, quantifies five classes of node/edge inconsistency against an independent reference graph, and exports schema-valid OSCAL SSP and AR artifacts.
In: 2nd IEEE/ACM International Conference on AI-powered Software, AIware 2025, Seoul, Republic of Korea, November 19-20
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
CONDITIONAL 5representative citing papers
Constrained decoding derived from OpenAPI specifications eliminates hallucinated web API calls in LLM-generated code and substantially improves correctness across 24 models, while retrieval-augmented generation yields inconsistent, model-dependent gains.
LLMs achieve only modest understanding of HMSC formal semantics at 52 percent accuracy, performing strongly on basic constructs but weakly on abstractions and traces.
REAP automatically filters real developer-AI sessions into an executable coding benchmark, Harvest, that separates five frontier models' solve rates from 42.9% to 58.2%.
An in-IDE multi-agent tutor called Prompt Coach uses Socratic guidance to improve developers' code-generation prompts, showing a 13.7% quality gain in a 15-person pre/post study.
citing papers explorer
-
Reverse Engineering Compliance: A Dual-Graph Verification Framework for Auditing Legacy IT Security Concepts
ASSERT extracts legacy IT security concepts into document graphs, quantifies five classes of node/edge inconsistency against an independent reference graph, and exports schema-valid OSCAL SSP and AR artifacts.
-
Mitigating Errors in LLM-Generated Web API Invocations via Retrieval-Augmented Generation and Constrained Decoding
Constrained decoding derived from OpenAPI specifications eliminates hallucinated web API calls in LLM-generated code and substantially improves correctness across 24 models, while retrieval-augmented generation yields inconsistent, model-dependent gains.
-
(How) Do Large Language Models Understand High-Level Message Sequence Charts?
LLMs achieve only modest understanding of HMSC formal semantics at 52 percent accuracy, performing strongly on basic constructs but weakly on abstractions and traces.
-
REAP: Automatic Curation of Coding Agent Benchmarks from Interactive Production Usage
REAP automatically filters real developer-AI sessions into an executable coding benchmark, Harvest, that separates five frontier models' solve rates from 42.9% to 58.2%.
-
Prompt Coach: An Empirical Evaluation of an Agentic Tutor for Learning Prompt Engineering in Software Development
An in-IDE multi-agent tutor called Prompt Coach uses Socratic guidance to improve developers' code-generation prompts, showing a 13.7% quality gain in a 15-person pre/post study.